Electronic data processing (EDP) equipment, such as servers, storage devices, or the like, are generally coupled to alternating current (AC) power sources in a data center and require very high reliability. For this reason, this equipment is generally coupled to one or more uninterruptible power sources (UPS). When redundant power sources (e.g., A and B power sources) are supplied in a data center, the data center manager must manage the provisioning and capacity demand for both of the sources. The provisioning must be done so that if either of the two sources fails, the remaining power source has sufficient power capacity to carry the total load of the equipment. However, the complexity of delivering power from a UPS to the equipment often creates numerous possibilities for interruption. For example, interim circuit breakers, power connection plugs and receptacles, power distribution units (PDUs), power strips, and other distribution systems are often placed in the circuit path between large UPS systems and the equipment. These components increase the probability of an interruption or disconnection of the equipment from the power sources.
Alternatively, some EDP equipment include only one power supply and one AC power input. In this configuration, the equipment is subject to the failure of the single AC source. EDP equipment may contain a dual power supply arrangement that can provide direct current (DC) power to the internal circuits of the equipment from two separate AC sources. In this arrangement, the failure of one of the AC sources will result in the equipment load being supplied from the alternate DC power supply in the unit. At times when both AC sources are present, the load is either shared by both power supplies, or favored to one of the power supplies. These systems, sometimes referred to as “redundant supplied” systems, may be a final line of defense for reliable power delivery to the electronic circuits within the equipment. However, these solutions may be costly due to the additional power supplies that may be required. In addition, the added components generate more heat, which may be extremely undesirable in many applications. Further, the additional components may require rack space, which may be extremely costly. Finally, the additional components may increase the number of cords (e.g., network cords, power cords, and the like), which may significantly reduce the cooling airflow in the racks, thereby reducing the overall maximum achievable deployment density and increasing energy costs.
It is against this background that the automatic transfer switch module described herein has been developed.